Local Government Digital Policy and AI Marketing Innovation: A Multi-level Moderated Mediation Analysis of China's Fresh Corn Industry

Abstract

In the context of agricultural digital transformation, how local government digitalization policies influence enterprise technological innovation has become a critical theoretical and practical issue. This study constructs a multi-level moderated mediation model based on the Technology-Organization-Environment (TOE) framework and Innovation Diffusion Theory to explore how local government digitalization policies affect enterprise AI marketing strategy adoption and market performance in China's fresh corn industry. Using survey data from 128 fresh corn enterprises across 15 provinces and 3,842 consumer questionnaires, we employ Partial Least Squares Structural Equation Modeling (PLS-SEM) for empirical analysis.The findings reveal that: (1) local government digitalization policies significantly promote enterprise AI marketing adoption (β=0.21, p<0.01), which positively impacts market performance (β=0.34, p<0.001), with the model explaining 58.7% of variance (R²=0.587); (2) policy environment characteristics significantly moderate the policy-enterprise relationship, with effects in high-support environments (0.43) being 2.5 times those in low-support environments (0.17); (3) industry competition intensity strengthens the AI adoption-performance relationship (β=0.21, p<0.01); (4) organizational learning capability (indirect effect=0.126) and customer satisfaction (indirect effect=0.187) play partial mediating roles; (5) regional heterogeneity analysis shows policy effects in eastern regions (β=0.38) significantly exceed western regions (β=0.19).This study extends digital governance theory applications in agriculture and provides empirical evidence for local governments to formulate differentiated digital agriculture support policies and for enterprise technological innovation decision-making.